J. Org. Chem. 1992,57, 1610-1613
1610
(racemicsulfoxide); mp 74.5 “C(pure S enantiomer); [ a I n = ~ -54’ ( C 1.98, CHCl3); ’H NMR (CDC13) 6 2.40 ( 8 , 3 H), 2.69 (8, 3 H), 7.28-7.55 (AA’BB’m, 4 H); ‘H NMR analysis with the chiral shift reagent, (S)-(+)-2,2,2-trifluoro-l-(9-anthryl)ethaol indicated the sulfoxide to be enriched in the S enantiomer (Le., 30% ee).
Preparation of 1-Chloroalkyl Hydroperoxides by the Addition of Hydrogen Chloride to Carbonyl Oxides
Walter V. Turner* and Siegmar Ghb GSF-Znstitut fur Okologische Chemie, Schulstrasse 10, 8050 Freising-Attaching, Federal Republic of Germany Received September 4, 1991
Carbonyl oxides, formed as intermediates in ozonolysis, are well-known to react with “participating solvent8” such as water, alcohols, hydrogen peroxide, alkyl hydroperoxides, carboxylic acids, ammonia, amines, and hydrogen cyanide to yield a-heteroatom-substituted hydroperoxides? We have previously made 1,l-dichloroalkylhydroperoxides by the addition of anhydrous hydrogen chloride to achlorinated carbonyl ~xides.~BWe report here the first examples of 1-chloroalkylhydroperoxides, prepared by the analogous addition of HC1 to carbonyl oxides that do not already have an a-chlorine atom. The lack of selectivity in the cleavage of the primary ozonide from unsymmetrical olefins and the formation of a highly reactive aldehyde or ketone alongside the desired carbonyl oxide would be disadvantageous in attempts to obtain the title compounds selectively. We prepared the 1,l-dichloroalkyl hydroperoxides from symmetrical double-bond-dichlorinated olefins; the byproduct, an acid chloride, reacted slowly enough with the hydroperoxides that it could be stripped out of the mixture after the reaction. In general it is possible to obtain carbonyl oxides selectively from unsymmetrical olefins when one doublebond carbon has an electron-withdrawing substituent such as a halogen or an alkoxy group (eq 1). If there is a
-
R1R2C+OO-+ R3COC1 (1) 1 chlorine atom on C-1 of a terminal olefin, there is the additional advantage that the acid chloride is formyl chloride, which rapidly decomposes into carbon monoxide and hydrogen chloride. This is the strategy that we have used in the present work to examine the reaction of HC1 with a number of carbonyl oxides, most of which have no electronegative substituent in the a position. In Tables I and I1 we show the olefins that we have subjected to ozonolysis in methyl formate in the presence of HC1 and the carbonyl oxides presumed to be intermediates in the reactions. Table I has the carbonyl oxides that were found to add HC1, and Table I1 those that did not do so. Carbonyl oxides la-e are clearly formed under these conditions and readily add HCl (eq 2). R1R2C=CC1R3+ O3
+ HC1-
R1R2CC100H (2) 2 Hydroperoxide 2b proved to be stable enough to be distilled and to give fair elemental analyses. This stability 1
(1) Bailey, P. S. Ozonation in Organic Chemistry; Academic Press: New York, 1978 Vol. I, p 111-118. (2) Gab, S.; Turner, W. V. J. Org. Chem. 1984,49, 2711-2714. (3) Gab, S.; Turner, W. V.; Hellpointner, E.; Korte, F. Chem. Ber. 1985, 118, 2571-2578.
is comparable with that of the analogous a-methoxy hydroperoxide C1CH2CH(OCH3)00H,which has been described by others? The @-brominatedhydroperoxide 2c was not distilled, but it was similarly stable. The methoxy and acetoxy analogues 2d and 2e were more labile; thus, although some NMR samples of 2d were unchanged after 1 h at 35 “C, others decomposed with the evolution of considerable heat when allowed to warm above 0 “C. The lower part of Table I shows some chlorinated hydroperoxide that have previously been reported (2g-i) and one (20 made in this study, in all of which the carbonyl oxide already has an a-chlorine substituent. The yield of trichloromethyl hydroperoxide (2g) is unknown; since the ozonolysis of C2C14is very slow, the alternative preparation of 2g by light-induced oxidation of chloroform is preferable? Nevertheless, the known ester CC1300COCH3was prepared in detectible amounts by prolonged treatment of C&14 with ozone in the presence of HC1 and acetyl chloride; it was identified by comparison with authentic material.5 Except for the simplest carbonyl oxide, la, all those so far found to add HC1 have electron-withdrawing groups in either the a or the j3 position. That this is not a sufficient criterion, however, is attested by the failure of carbonyl oxide lm to produce CGH5CCl200H.It is uncertain just how the failure occurs in this example: in the formation of lm, in the addition of HC1, or in some instability of the hydroperoxide once formed. In some of the other cases, the course of the reaction was clearer. Thus, ozonolysis of 1-methoxycyclopentenegave the same products (polymers) with HC1 as withoute6 This appeared also to be the case for j3-bromostyrene (benzaldehyde and diphenyltetroxane as the main products). Ozonolysis of two olefins that are expected to yield the carbonyl oxide (CH3)2C+OO-(lk) failed to give any evidence for the formation of (CHJ2CC100H (2k). Instead, the major product was acetone peroxide (the cyclic dimer or trimer of lk). In this context it must be mentioned that other workers, carrying out the ozonolysis of 2-methyl-3chloro-2-butene on the surface of polyethylene, found evidence from NMR spectra and product studies for the formation of the acetate ester of hydroperoxide 2k.’ It was suggested that the ester arose by rearrangement of the normal ozonide; in contrast to the esters of the 1,l-dichloroakyl hydroperoxides we have previously described, however, it was both thermally and chemically unstable, reacting spontaneously with either water or silica gel. If hydroperoxide 2k is equally sensitive, it may well not have survived our reaction conditions. The only a-chlorinated hydroperoxides that we have been able to acetylate with acetyl chloride are 2g-i, and all of the esters have been described p r e v i o u ~ l y . ~The ,~~~~~ failure of the others to yield acetates cannot in all cases be due to instability of the hydroperoxides: 2b, for example, is stable for days in acetyl chloride at 4 “C. This resistance toward acetylation with acetyl chloride is particularly surprising in light of the fact that 1,l-dichloroethyl hydroperoxide was readily acetylated in good yield when allowed to stand in acetyl ~ h l o r i d e . ~ These results led us to question the assigned structures, but the spectroscopic evidence for them is strong (the (4) (a) Tempesti, E.; Fornaroli, M.; Giuffre, L.; Montoneri, E.; Airoldi, G. J. Chem. SOC.,Perkin Trans. I 1983,1319-1323. (b) Griesbaum, K.; Meister, M. Chem. Ber. 1985,118,845-850. (5) Gab, S.;Turner, W. V. Angew. Chem. 1985, 97,48. (6) Gab, S.; Piesch, P.; Turner, W. V.; Korte, F., to be published. (7) Griesbaum, K.; Greinert, R. Chem. Ber. 1990, 123, 391-397. (8) Griesbaum, K.; Hoffman, P. J. Am. Chem. SOC.1976, 98,
2877-2881.
0022-326319211957-1610$03.00/00 1992 American Chemical Society
Notes
J. Org. Chem., Vol. 57, No. 5, 1992 1611 Table I. Carbonyl Oxides Giving Hydrowroxides with HCl starting olefin carbonyl oxide hydroperoxide CH24HCl la, CH200 2a, C1CH200H ClCHzCH4HCl lb, ClCHzCHOO 2b, ClCHZCHClOOH BrCH2CH=CHC1 IC, BrCHICHOO 2c, BrCH2CHC100H CHsOCH&H=CHCl Id, CH30CHzCH00 2d, CH30CHzCHC100H CH3COOCHZCH=CHCl le, CH3COOCHzCH00 2e, CH3COOCH2CHC100H CHCl=CHCl If, ClCHOO 2f, CLCHOOH ccl*=Ccl~ le, Cl~COO 2g, CGCOOH CH3CCl=CClCH3 lh, CH3CC100 2h, CH3CC1200H C2H&Cl=CClC2HS li, C2H5CC100 2i, CzH5CC1200H
Table 11. Carbonyl Oxides Not Giving Hydroperoxides with HCl starting olefin presumed carbonyl oxide CH&H=CHCl lj, CH3CHO0 (CH3)2C=CHC1 lk, (CH3)2COO (CHJ&=CClCH3 lk, (CH3)ZCOO C6H5CH4HBr 11, C6H5CHOO C6H5CC1=CC1C6H5 Im, C,&CC100 ~-CH~O-C-C~H, In, CH30CO(CH2)3CH00
infrared spectrum of 2b, for example, has an OH-stretch absorption at 3510 cm-', and in most cases the OH proton was visible in the 'H NMR spectrum); we have also been able to confirm the presence of the hydroperoxide group by another derivatization. Apparently the reaction of the hydroperoxides 2a-e with acetyl chloride is very slow, and any ester formed is too unstable to be detected. Thus, introduction of gaseous formaldehyde into a solution of the hydroperoxide 2b leads to the formation of 3b, a rather stable substance. Analogous addition products were prepared from 2a and 2c-f (eq 3). The addition of aldehydes to hydroperoxides is an established reaction, and in a number of cases the products are at least as stable as the hydroperoxides themselve~.~*~ 2 + HCHO R1R2CC100CH,0H (3) 3 3 + AcCl R1R2CC100CH20Ac HC1 (4) 4 There are a few reports in the literature of the esterification of such addition products.1° Treatment of compounds 3a-d,f with acetyl chloride yields the acetate esters la-d,f (eq 4). Acetyl chloride reacted with 3g to give the ester of 2g, not 3g. One product from the attempted acetylation of 3e appeared from the NMR spectra to be the expected ester 48, but after distillation it was found to contain only trace amounts of chlorine. Attempts to increase the yields of the esters by adding bases led to decomposition of the peroxides. This is expected, since the hydroperoxidealdehyde addition products are known to be unstable toward base." The 'H and 13CNMR spectra of compounds 2b-e, 3b-e, and 4b-d form a self-consistent set. The proton of the CHCl group resonates at ca. 6 6.1 and is coupled differently to the diastereotopic protons of the neighboring CH2group; it is little affected by the addition of formaldehyde. The carbon of the CHCl group falls in the narrow range 6 93.6-96.5 in 2b-e and experiences a small y shift in 3b-e and 4b-d. Most distinctive is l J C H of this carbon, which is consistently 178-183 Hz. We are not aware of any 13C
-
-
+
(9) Cf. Marklund, S. Acta Chem. Scand. 1971,25, 3517-3531. (10)(a) Griesbaum, K.; Keul, H.; Agarwal, S.; Zwick, G. Chem. Ber. 1983,116,409-415. (b) Griesbaum, K.; Zwick, G. Chem. Ber. 1985,118, 3041-3057. Griesbaum, K.; Zwick, G. Chem. Ber. 1986, 119, 229-243. (11) (a) Petrovskaya, G. A.j Vilenskaya, M. R.; Karpenko, A. N.; Panchennko, Y. V. Zh. Org. K h m . (Eng.)1979,16,829-831. (b) Mizyuk, Y. L.; Mamchur, L. P.; Vilenskaya, M. R.; Yurzhenko, T. I. Ukrain, Khim. Zh. 1972,38, 795-798.
NMR spectra in the literature for formaldehyde-addition products of hydroperoxides. The OCH20 carbon at 6 92.4-93.0 falls near the range of the CHCl carbons in 3b-e, but it is readily distinguished by both ita multiplicity and (166167 Hz); on esterification it experiences ita lower lJCH a small upfield shift, and its coupling constant increases to 171 Hz. The protons of the OCH20group exhibit the expected downfield shift (ca. 0.5 ppm) on esterification. These chemical-shift and coupling-constant effects are almost identical with those shown by the mono- and diacetates (6 and 7) of bis(hydroxymethy1) peroxide (5), which were made for comparison. The resonance of the CHC12proton in 2f may easily be overlooked, since it is nearly (sometimes exactly) the same as that of chloroform. The lJcH value of 205 Hz in 2f is also remarkably close to that of chloroform at 208 Hz.12 Experimental Section All chemicals and solvents were from Merck, Darmstadt, Federal Republic of Germany. 1,3-Dichloropropene (DCP) was a mixture of the cis and trans isomers. Ozone was generated with a Fischer Model 502 generator, and the mixture of HC1 and O3 gases was made by adding HC1 from a lecture bottle to the O3 stream through a T-connector. Elemental compositions and molecular weights (MW) by vapor pressure osmometry (VPO) were determined by the laboratory of F. Pascher, RemagenBandorf, Federal Republic of Germany. Molecular weights by cryoscopy (cry) in benzene were determined with an apparatus from Knauer, Oberursel, Federal Republic of Germany. NMR spectra were made in CDC1, solution on a Varian CFT-20 ( W ) , its 80-MHz modification ('H) or a Perkin-Elmer R32 (90-MHz lH).IR spectra were made of CCl, solutions on a Perkin-Elmer Model 577 spectrometer. GCMS spectra were made on a Hewlett-Packard Model 5995A. 1-Chloro-3-methoxypropene(ca. 1:l cis/trans mixture) was prepared by methanolysis of DCP.13 1-Chloro-3-acetoxypropene (cis/trans) was made by treating DCP with NaOAc, not in ethanol, as in the literature,', but in AcOH. l-Chlor0-3-hydroxypropene'~ (&/trans) was prepared by basic hydrolysis of the acetate. 1Chloro-3-bromopropene (cis/trans) was prepared from the alcohol by a procedure described for deny1 alcohols.'6 The NMFt spectra indicated up to 25% of the starting alcohol as an impurity. 1,2-Dichloroethyl Hydroperoxide (2b). A solution of 5.0 g (45.1 mmol) of DCP in 100 mL of dry methyl formate was saturated with anhydrous hydrogen chloride at -10 "C and treated with a stream of ozone and hydrogen chloride until it became blue (ca. 75 min). Removal of solvent under vacuum at 0 "C left an oil that could be distilled (25 "C, 0.3 Torr) to yield 2.51 g (42.5%). Active oxygen (iodometric in acetic acid): 11.25%, calcd 12.22%. Elemental Anal. Calcd for CzH4Cl2O2(MW 130.95): C, 18.34; H, 3.08; 0, 24.43; C1, 54.14. Found C, 18.22; H, 3.54; 0, 25.6; C1,50.3. M W found (cry) 142; (VPO)269. 'H N M R 6 9.0 (br, OH), 6.09 (dd, J = 8.1, 4.2 Hz, CHCl), 3.85 (dd, J = 11.5, 8.1 Hz, (12) Stothers, J. B. Carbon-13 NMR Spectroscopy; Academic.Press: New York, 1972; p 326. (13) Kirrmann, A.; Pacaud, A.; Dosque, H. Bull. SOC. Chim. Fr. 1934, I (5), 860-871. (14) Martinoff, A. Chem. Ber. 1875, 8, 1318-1319. (15) van Romburgh, M. P. Bull. SOC.Chim. Fr. 1881,36 ( Z ) , 549-555. (16) Black, D. K.; Landor, S. R.; Patel, A. N.; Whiter, P. F. J. Chem.
SOC.C 1967, 2260-2262.
1612
Notes
J. Org. Chem., Vol. 57, No. 5, 1992
H - ~ L Y3.72 ) , (dd, J = 11.5, 4.2 Hz, H-26). 13CNMR 96.3 (JCH = 181 Hz, CHCl), 43.6 ( J C H = 156 Hz, CHzCl). IR: strong absorptions at 3510,1430,1355,1182,1100,1042, and 700 cm-'. The other hydroperoxides were prepared similarly, but at -20 "C. Chloromethyl Hydroperoxide (2a). 'H NMR: 6 7.5 (br 8 , OH), 5.80 (8, CHZC1). "C NMR 6 85.7 (JCH= 176 Hz). 1-Chloro-2-bromoethyl Hydroperoxide (2c). 'H NMR 6 8.9 (br 8, OH), 6.08 (dd, J = 8.1,4.2 Hz,CHCl), 3.85 (dd, J = 11.5, 8.1 Hz, H - ~ L Y 3.69 ) , (dd, J = 11.5, 4.3 Hz, H-28). "C NMR 6 95.8 (JcH 182 Hz, CHCl), 30.3 (JCH= 157 Hz, CHzBr). 1-Chloro-2-methoxyethylHydroperoxide (2d). 'H NMR 6 8.5 (br, OH), 6.08 (dd, J = 6.6,4.7 Hz, CHCl), 3.80 (dd, J = 11.8, 6.8 Hz, H-2a),3.76 (dd, J = 11.7,4.7 Hz, H-ZB), 3.45 (8, OCH3). 13C NMR 6 96.2 ( J C H 180 Hz, CHCl), 72.7 (JCH 151 Hz, MeOCH,), 59.7 (JCH= 143 Hz, OCH3). 1-Chloro-2-acetoxyethyl Hydroperoxide (Ze). 'H NMR: 6 8.5 (bra, OH), 6.05 (dd, J = 7.8,4.0 Hz,CHCl), 4.57 (dd, J = 12.3, 7.8 Hz, H-2a),4.34 (dd, J = 12.4,4.0 Hz, H-26), 2.13 (CH3). 13C NMR 6 171.6 (C--O), 95.0 (JCH= 178 Hz, CHCl), 63.9 (JCH= 153 Hz, AcOCH~),20.6 (JCH 130 Hz, CH3). Dichloromethyl Hydroperoxide (2f). 'H NMR 6 9.0 (br S, OH), 7.27 (8, CHC1.J. "C NMR 6 99.5 (JCH 205 Hz). l&Dichloroethyl Hydroxymethyl Peroxide (3b). Hydroperoxide 2b was prepared as above from 1.0 g of technical DCP, and the solvent was removed under vacuum. The oil was redissolved in 25 mL of CHzClzand cooled to 0 "C. Formaldehyde was generated by slowly warming 0.5 g of paraformaldehyde in a small flask with a flame. A stream of N2 was passed through this flask and into the solution of hydroperoxide until all the paraformaldehyde had volatilized. This took about half an hour. After another half hour a t 0 "C, the solvent was removed under NMR spectrum the purity of the product vacuum. From a was estimated at ca.80%. It could be chromatographed on silica gel with 20% ethyl acetate in CHC13, but the purity was not improved. On attempted distillation under vacuum, the product decomposed with evolution of a gas. 'H NMR 6 6.1 (dd, J = 3-4,6-7 Hz, CHCl), 5.28 (a, OCHzO),4.5 (br, OH), 3.6-4.1 (m, CHZCl). 13C NMR 6 95.1 (JCH= 183 Hz, CHCl), 93.1 (JCH= 167 Hz, OCH20),44.2 ( J C H = 157 Hz, CHzC1). IR. strong absorptions at 3600,1430,1180, and 700 cm-' as well as a group of absorptions between 950 and 1110 cm-'. The addition of formaldehydeto the other hydroperoxides was carried out at -20 "C. Chloromethyl Hydroxymethyl Hydroperoxide (3a). 'H NMR: 6 5.84 (t, J = 1.0 Hz, CHzCl), 5.24 (t,J 1.0 HZ,OCH2O), OH proton not observed. 13C NMR: 6 92.9 (JCH= 167 Hz, OCHZO), 83.6 (JCH 177 Hz, CH2Cl). 1-Chloro-2-bromoethyl Hydroxymethyl Peroxide (30). 'H NMR 6 6.15 (dd, J = 8.0, 4.5, CHCl), 5.28 (br s, J = 0.7 Hz, OCHzO),3.82 (dd, J = 11.5, 8.0 Hz, H-24, 3.70 (dd, J = 11.5, 4.6 Hz, H-28), OH proton not observed. 13CNMR: 6 94.1 ( J C ~ = 183 Hz, CHCl), 92.5 (JCH= 166 Hz, OCHZO), 30.7 (JCH= 157 Hz, CH2Br). 1-Chloro-2-methoxyethylhydroxymethyl peroxide (3d)was obtained as a mixture with 2d. 'H NMR: 6 8.35 (br s, OH), 6.18 (m, CHCl), 5.25 (d, J = 0.9 Hz, OCHzO),3.6-4.0 (m, MeOCHz), 3.46 ( 8 , CH30). '9C NMR: 6 94.3 (JCH= 180 Hz,CHCl), 92.4 (JCH = 166 Hz, OCHZO), 73.1 (JCH = 146 Hz, MeOCH2),59.5 (JcH = 143 Hz, CH30). 1-Chloro-2-acetoxyethyl Hydroxymethyl Hydroperoxide (36). '3C NMR: 6 170.9 (M). 93.6 (Jaw = 181Hz.CHCl). 93.0 Dichloromethyl hydroxymethyl hydroperoxide (3f) was not obtained pure; the NMR spectrum was poorly resolved. 'H NMR 6 7.31 (8, CHClZ),5.93 (br 8, OH), 5.34 (8, OCH20). Trichloromethyl Hydroxymethyl Peroxide (3g). 'H NMR: 6 5.45 (8, OCHzO),3.6 (br 8, OH). 13CNMR 6 113.7 (CC13),93.6 (JCH * 168 Hz, OCH2O). l,2-Dichloroet hyl Acetoxymet hyl Peroxide (4b). The peroxide 3b was prepared as above from 0.50 g of DCP, but not chromatographed. Stripped of solvent, it was redissolved in 5 mL of acetyl chloride and held at 7 "C for 3 h. The volatiles were removed at aRpirator vacuum, and the residual oil was chromatographed on silica gel with a 1:l mixture of hexane and CHCl,.
The combined peroxide fractions (0.56 g, 62%) were distilled in a miniature sublimator to yield a few drops for analyses. Anal. Calcd for C5H8C1204(MW 203.00): C, 29.58; H, 3.96; 0, 31.53; C1,34.93. Found: C, 29.24; H, 4.01; 0, 32.1; C1,34.3. MW (cry): found 197. 'H NMR 6 6.04 (dd, J = 8,4.5 Hz, CHCl), 5.72 (a, OCH20),3.90 (dd, J = 12,8 Hz, H-2a),3.80 (dd, J = 12,4.5 Hz, H-26), 2.13 (8, CH3). 13CNMR 6 169.8 (C=O), 94.7 (JCH= 182 Hz,CHCl), 89.8 (Jm = 171 Hz,OCHZO), 43.9 (JCH= 154, CHZCl), 20.9 (JcH= 130 Hz, CH3). I R strong absorptions at 1770,1210, and lOO(k1035 cm-l. GCMS m/e 107/109,77/79 (base peak), 73, and 49/51. The other esters were prepared similarly. Chloromethyl Acetoxymethyl Peroxide (48). Anal. Calcd for C4H7C104(MW 154.55): C, 31.08; H, 4.57; 0,41.41; C1,22.94. Found C, 31.01; H, 4.59; 0,41.3; C1, 22.9. MW (cry) 164; ( W O ) 180. 'H NMR: 6 5.78 (t, J = 1.0 Hz, ClCHJ, 5.69 (t, J = 1.0 Hz, OCHZO), 2.15 (8, CH3). "C NMR 6 169.9 ( C d ) , 90.0 (JCH= 171 Hz, OCHZO), 83.1 (JCH 176, ClCHz), 20.9 ( J C H = 130 Hz, CH3). IR: strong absorptions at 1760,1210,and 1010 cm-' and moderate ones at 1365, 1305 and 685 cm-'. l-Chloro-2-bromoethyl Acetoxymethyl Peroxide (4c). Anal. Calcd for C5H8ClBr04(MW 247.44): C, 24.27; H, 3.25; 0,215.86; C1, 14.33; Br, 32.29. Found: C, 24.08; H, 3.28; 0, 25.9; C1, 14.4; Br, 32.2; MW (cry) 250. 'H NMR 6 6.09 (ddt, J = 8.1,4.6,0.8 Hz, CHCl), 5.72 (t, J = 0.8 Hz, OCH,O), 3.75 (dd, J = 11.5, 8.1 Hz, H - ~ L Y3.70 ) , (dd, J = 11.5,4.6 Hz, H-281, 2.15 (8, CH3). 13C NMR 6 169.7 ( C d ) , 94.3 (JCH= 183 Hz, CHCl), 89.8 (JCH= 171 Hz, OCHZO), 30.4 ( J C H = 157 Hz, BrCHz), 20.9 (JCH= 130 Hz, CH3). I R strong absorptions at 1765,1207, and lo00 cm-'. 1-Chloro-2-methoxyethyl Acetoxymethyl Peroxide (4d). Anal. Calcd for C6Hl1C1OS(MW 198.61): c , 36.28; H, 5.58; 0, 40.28; C1,17.85. Found: C, 36.30; H, 5.74; 0,41.1, C1, 17.2; MW (cry) 198. lH NMR: 6 6.09 (ddt, J = 6.5,5.3,0.6 Hz, CHCl), 5.71 (t,J = 0.5 Hz, OCH,O), 3.73 (m, H - 2 ~ 43.42 , (s, CH30), 2.13 (8, CH3). '3c NMR: 6 169.9 (c=O),95.0 (JCH = 180 Hz,CHCl), 89.9 (JcH = 171 Hz, OCHZO), 73.5 (JCH= 146 Hz, OCHZC), 59.6 (JCH 140 Hz, CH30), 20.9 (JCH= 126, CHd. Dichloromethyl Acetoxymethyl Peroxide (41) (containing AcOH and unidentified products). 'H NMR 6 7.29 (t, J = 1.1 HZ,CHClJ, 5.77 (d, J = 1.1Hz,OCHZO), 2.15 (8, CH3. '3C NMR: 6 169.7 (c=O), 97.4 (JcH = 205 Hz,CHClZ), 90.0 (JCH= 171 Hz, OCHZO), 20.7 or 20.8 (JCH= 130 Hz, CH3). Caution. Bis(hydroxymethy1) peroxide is reported to be extremely expl~sive,'~ and the acetate ester of trichloromethyl hydroperoxide has caused several explosions in our laboratory. The new compounds reported here should be assumed to be both explosive and toxic.
Acknowledgment. We are grateful to the Deutsche Forschungsgemeinschaft for financial support of this work. Registry No. la, 78894-19-6; lb, 13843404-5; IC,13843405-6; Id, 13843406-7; le, 138434-07-8; lf, 138434-08-9; lj, 68941-70-8; lk, 65339-02-8; 11,138458-85-2;lm, 138458-86-3; In, 138434-17-0; 2a, 138434-09-0;2b, 138434-10-3;2 ~138434-11-4; , 26,138434-12-5; 20,138434-13-6; 2f, 138434-147;2g, 94089-33-5; 38,40323-32-8; 3b, 138434-18-1; 3d,138434-20-5;3e,138434-21-6; 3f, 138434-22-7; 3g,138434-23-8;h,138434-249;4b, 138434-250; 4c, 138434-26-1; 4d,138434-27-2;4,13843428-3; 6,138434-29-4; 7,22721-77-3; 8a, 138434-30-7;Sf, 138434-31-8; 9,764-41-0; 10,13843432-9;DCP, 542-75-6; CHz=CHCl, 75-01-4; BrCH2CH=CHCl, 3737-00-6; CHSOCH&H=CHCl, 2806-79-3; CH,COOCH,CH=CHCl, 13042-00-7; CHCl=CHCl, 540-59-0; CH&H=CHCl, 590-21-6; (CH,),C=CHCl, 513-37-1; (CH&zC=CClCH3, 17773-65-8; CgH,CH=CHBr, 103-64-0; C6H5CC1=CClC6H5, 13700-82-8; 1C H ~ O - C - C ~1072-59-9. H~, Supplementary Material Available: 'H and 13C NMR resonances of 1-chloro-3-methoxypropene, 1-chloro-3-acetoxypropene, l-chloro-3-hydroxypropene, and 1-chloro-3-bromopropene; tabulated long-range '%-'H coupling constantsfor 2a,b, 3a,b, 4a,b; preparation and tabulated NMR spectra of 6 and 7; a discussion of the formation and NMR spectra of chloromethyl hydroperoxymethyl peroxide (sa)and dichloromethyl chloro(17) Wieland, H.; Sutter, H.Chem. Ber. 1930,63,66.
J. Org. Chem. 1992,57, 1613-1615
1613
hydroperoxymethylperoxide (8f); a discussion of the ozonolysis of 1,4-dichloro-2-butene(9) in the presence of HC1 to yield 1,2dichloroethyl l'-hydroxy-2'-chlorothyl peroxide (10); 'HNMR spectra of 2a + 8a, 2b-e, 2f, 3a-d,f,g,4a,b,d, and 6; 13C NMR spectra of 2a + 8a, 2b-f, 3a-d,g, 4a,b,d,f,and 6; and IR spectra of 2b, 3b, and 4b (42 pages). Ordering information is given on any current masthead page.
Practical Synthesis of a n Enantiomerically P u r e Intermediate of the Lactone Moiety of Mevinic Acids Serafh Valverde,* J. CrisMbal Lbpez, Ana M. Gbmez, and Silvestre Garcia-Ochoa Znstituto de Qulmica Orgdnica General, (CSZC), Juan de la Cierva 3, 28006 Madrid, Spain
Chart I
no
1 a; R d
(+)compactln
b; R-Me (+)-mevlnolln
HO
Received September 10,1991
Since the recognition of hypercholesterolemia as a risk factor of atherosclerosis and coronary heart disease,l there have been multiple efforts to identify chemical entities capable of regulating the plasma level of this steroL2 Mevinic acids3are part of a family of fungal metabolites that have attracted considerable attention due to their biological activities as inhibitors of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase, the rate-limiting enzyme in cholesterogenesis in man1.4~~Compactins (la) and mevin01in~~J (lb) (Chart I) are two prominent members of this family of clinically useful hypocholesterolemic agents. The key structural feature common to all mevinic acids3is the 8-hydroxyvalerolactone function A which, in its open form, closely mimics mevalonic acid, a crucial intermediate in the terpenoid biosynthetic pathway leading to cholesterol. Structure-activity relationships in previous series revealed8 that the chiral lactone moiety, A, is essential for strong biological activity, whereas the hexahydronaphthalene half allows more structural variations. Recent reportas on the discovery of active analogues, B, which retain the ladone portion of mevinic acids have increased the potential interest of enantiospecific routes to suitable chirons.lo
In view of the nature of the lactone portion, a number of syntheses involved starting materials derived from carbohydrates." Shorter chains derived from chiral, nonracemic hydroxy acids and similar sources have also been uaed.12 The lactone moiety has also been constxuct8d by employing a hetero-Diels-Alder approach, a protocol extensively investigated by Danishefsky and co-worker~.'~ A practical approach to incorporate the key chiral hydroxy lactone portion A in the synthesis of the natural compounds l a , l b or the analogues B is to employ the masked lactols as precursors of electrophilic species. Consequently, the synthesis of multigram amounts of methyl 2,4-dideoxy-3,6-bis(0-p-nitrobenzoy1)-D-erythrohexopyranosides 2 from D-glucose using a few simple and high yielding steps is of considerable interest. 2,35,6-di-O-isopropylidene-~-glucose diethyl dithioacetal (3) is readily available from D-glucose using standard methods.14J5 Compound 3 (Scheme I) was transformed
(1) Anderson, K. M.; Castelli, W. P.; Levy,D. J. Am. Med. Assoc.1987, 257,2176. Steinberg, D. Circulation 1987, 76, 502. (2) Havel, R. J. J. Clin. Invest. 1988,81, 1653. (3) For a comprehensive review of mevinic acid syntheses, see: Rosen, T.; Heathcock, C. H. Tetrahedron 1986,42,4909. (4) (a) Kaneko, I.; Hazama-Shimada, Y.; Endo, A. Eur. J.Biochem. 1978,87,33. (b) Alberts, A. W.; Chen, J.; Kuron, G.; Hunt, V.; Huff, J.; Hoffmann, C.; Rothrock, J.; Lopez, M.; Joshua, H.; Harris, E.; Patchett, A.; Monaghan, R.; Currie, S.;Stapley, E.; Albers-Schomberg, G. A.; Hensene, 0.; Hirshfield, J.; Hoogsteen, K.; Liesch, J.; Springer, J. Proc. Natl. Acad. Sci. U.S.A. 1980, 77, 3957. (5) For reviews, see: Endo, A. J. Med. Chem. 1985,28,401. Vega, L.; Grundy, S. J. Am. Med. Assoc. 1987,257,33. Kathawala, F. G. Med. Res. Reu. 1991, 11, 121. (6) (a) Endo, A.; Kuroda, M.; Tsujita, Y. J. Antibiot. 1976,26, 1346. (b) Brown, A. G.; Smale, T. C.; King, T. J.; Hasenkamp, R.; Thompson, R. H. J. Chem. SOC.. Perkin Trans 1 1976. 1165. (7) Endo, A. J. Ahtibiot. 1979,32, 854. ' (8) See among others: Jendrda, H.; Baader, E.; Bartmann, W.; Beck, G.; Bergmann, A.; Granzer, E.; Kerekjarto, B. v.; Kesseler, K.; Kraue, R.; Schubert, W.; Wees, G. J. Med. Chem. 1990,33, 61. (9) (a) Stokker, G. E.; Alberta, A. W.; Deana, A. A.; Gilfillan, J. L.; Huff, J. W.; Smith, R. L. J. Med. Chem. 1986,29,852. (b) Sandoz. Eur. Appl. EP-A-0221025,1987. (c) Sliekovic, D. R.; Picard, J. A.; Roark, W. H.; Roth, B. D.; Ferguson, E.; Krauee, B. R.; Newton, R. S.; Sekerke, C.; Shaw, M. K. J. Med. Chem. 1991,34,367. (d) Sit, S. Y.; Parker, R. A.; Motoc, I; Han, W.; Balasubramanian,N.; Catt, J. D.; Brown, P. J.; Harte, W. E.; Thompson, M. D.; Wright, J. J. J. Med. Chem. 1990,33,2982. (e) Beck, G.; Keeeeler, K.; Baader, E.; Bartmann, W.; Granzer, E.; Jendralla, H.; v. Kerekjarto, B.; Kraue, R.; Paulus, E.; Schubert, W.; Wess, G . J. Med. Chem. 1990, 33, 52. And previous references by these research groups therein.
(10) Hanessian, S. Total Synthesis of Natural Products: The Chiron Approach; Baldwin, J. E.; Ed.; Pergamon Press: Oxford, 1983. Hanessian, S. Aldrichimica Acta 1989, 22, 3. (11) (a) Prugh, J. D.; Deana, A. A. Tetrahedron Lett. 1982,23, 281. (b) Yang, Y. L.; Falck, J. R. Tetrahedron Lett. 1982,23,4305. (c) Denishefsky, s.;Kobayashi, s.;Kerwin, J. F., Jr. J. Org. Chem. 1982,47,1983. (d) Ho, P. T.; Chung, S. Carbohydr. Res. 1984,225,318. (e) Lee, T. J. Tetrahedron Lett. 1985,29,1255. (0Prugh, J. D.; Rooney, C. S.; Deana, A. A; Ramjit, H.G. J. Org. Chem. 1986,51,648. (9) Roark, W. M.; Roth, B. D. Tetrahedron Lett. 1988, 29, 1255. (h) David, D.; Gesson, J. P.; Jacquesy, J. C. Tetrahedron Lett. 1989,30, 6015. (12) (a) Majewski, M.; Clive, D. L. J.; Anderson, P. C. Tetrahedron Lett. 1984,25, 2101. (b) Rosen, T.; Taschner, M. 3.; Heathcock, C. H. J. Org. Chem. 1984,49,3994. (c) Guindon, Y.; Yoakim, C.; Berstein, M. A. Tetrahedron Lett. 1985,29,1185. (d) Baader, E.; Bartmann, W.; Beck, G.; Bergmann, A.; Fehlhaber, H.W.; Jendralla, H.;Keasler, K.; Saric, R.; Schiissler, H.; Teetz, V.; Weber, M.; Wess, G. Tetrahedron Lett. 1988, 29, 2563. (e) Davidson, A. H.; Floyd, C. D.; Lewis, C. N.; Myers, P. L. J. Chem. SOC., Chem. Commun. 1988,1417. (0 Bennet, F.; Knight, D. W.; Fenton, G. J. Chem. SOC.,Perkins Trans 1 1991,133. (13) (a) Danishefsky, S. J.; Simoneau, B. Pure Appl. Chem. 1988.60, 1555. (b) Danishefsky, S. J.; Kerwin, J. F., Jr.; Kobayashi, S. J. Am. Chem. SOC. 1982,104,358. (c) Wovkulich, P. M.; Tang, P. C.; Chadha, N. K.; Batcho, A. D.; Barrish, J. C.; Uskokovic, M. R. J. Am. Chem. SOC. 1989,111,2596. (14) Compound 3 has been previously prepared (75% yield) by reaction of D-glucose diethyldithioacetal and 4 equiv of 2-methoxypropene (Grindley, T. B.; Wickramage, Ch. Carbohydr. Res. 1987,167,105). In our hands, the simplest manner to carry out this operation was to dissolve the crude dithioacetal15in acetone. Allowing the solution to stand for 5-6 h at rt affords the required diacetonide 3, certainly due to the presence of residual acid from the previous step acting as a catalyst.
'on
Ai B
0022-3263192f 1957-1613$03.00/0 0 1992 American Chemical Society
2
